INTRUSION DETECTION ON CLOUD USING HYBRID MACHINE LEARNING TECHNIQUES

Computer Science Engineering

Cloud computing delivers ubiquitous and pay-per-use services, and as a result of these features, it attracts more consumers to utilize its services. Despite the many benefits of cloud computing, there are a few drawbacks. There is a risk of cloud-based attacks that compromise security features such as confidentiality, availability, and integrity. In order to identify attacks and improve cloud security, security solutions are required for both cloud users and cloud service providers. Cloud computing's primary concern is security. In the cloud, there are a lot of incursions. In order to identify intrusions, a variety of methodologies have been devised. However, no one method can accurately categories all types of assaults. The hybrid machine learning-based Intrusion Detection System (IDS) proposed in this study is a blend of supervised and unsupervised machine learning algorithms. There is a categorization of machine learning and deep learning technologies that are used to identify harmful user assaults in the network. Various researchers' techniques to detecting suspicious activity in the NSL KDD data utilizing tools are highlighted. We've also organized the papers' publishing throughout the last 19 years by year of publication and database source. In the last part of the research, we run an experiment to identify assaults in the dataset NSL-KDD. To identify the assault in the dataset, a machine learning intrusion detection model is developed using effective classifiers.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.